Rohde & Schwarz Universal Radio Communication Tester CMU200: PSU problem

[richnormand]

Ok just a quick update. Replaced the TOP200YN with the new(er) TOP222YN. Fast and easy, so why not.

Same behaviour as before. Considering the TOP can handle about 25W max and is switching the full primary current, it reinforce my comment in the first post that it most likely only handles the standby 12V.  The person that fixed their PSU changing the TOP most likely did not have the 12V standby to start with. I just sent him/her a pm to confirm.

Back to the bench it goes.

[Ericmax]

Hello everybody.
I'm Eric from France.

I bought a CMU200 on E-bay UK that I received yesterday.
Very happy with my purchase, I unwrapped my package.
the unit appears in good condition, so I plugged it to see all its possibilities.
The unit has started and the screen displayed the boot with the message "" CMOS checksum error - Defaults loaded "
I decided to take pictures to indicate the problem to the vendor and ask if we had to change the battery.
Unfortunately I only had time to take two photos that CMU is off by itself. all lights are turned off before the Panel.
I unplugged the CMU, handled the power button and reconnected but no lights are on and the CMU looks dead.
Retried this morning always the same, no light and no reaction at CMU.

Do you think my problem is the PSU ?

[PartialDischarge]

At least an orange led should be on when you plug the unit. There is a 12V standby voltage generated by the PSU.
If not present, then yes, the psu is dead and most probably is the top200 which handles that voltage.

[Ericmax]

Thank you MasterTech.

It is the symptom, no LED lit when plugged
you know where is this component in the psu ?

[richnormand]

Thank you MasterTech.

It is the symptom, no LED lit when plugged
you know where is this component in the psu ?

ONE:

The TOP200 is on the board 2 that connect between the two main boards. as shown in the two photos here.
Unfortunately you have to take apart the whole psu to get at it. It provides the 12V standby supply. This shows as the amber light at the front on the on/off switch. This is a good time to also check the electrolitic capacitors and small inductors while you are in there. Near the psu connector on the motherboard is a series of LED that gives the status of each power rail.  You can have a quick look to see if the 12V stanby is on. If not you can measure it at the connector using the info in previous posts here.

The initial TOP200 was susceptible to line transient and at the suggestion of MasterTech I replaced it with a TOP222YN from Mouser.  Unfortunately, in my case, the psu issues are much deeper than just the standby supply. I hope you will be lucky and that this is the only issue you have.

Also check the two small fuses at the power-line entrance of the main board.

TWO:

The errors you mentioned in your previous post might be a weak Li bios backup battery. Mine was almost dead. It is a regular button battery as used in computers but getting at it requires removing the whole front module and screen. A real pain. See photo attached.

Best of luck with it and don't forget to have fun!

[Ericmax]

Hi Richnormand  and thank you for all that information. 

I opened the CMU to display the control LED power supplies.
Of course. no LED is lit but I noticed that the LED +12V STB flashes once per second, which suggests that it is this part that is down.
It's really unlucky, the device functioned only 2 minutes after unpacking, just time to see the boot and BIOS problem.
I am negotiating with the seller to have a functional PSU but I think this problem of PSU is the weak point of all CMUand CRTU-RU

Thank you all for your help.

[PartialDischarge]

Hi Richnormand  and thank you for all that information. 

I opened the CMU to display the control LED power supplies.
Of course. no LED is lit but I noticed that the LED +12V STB flashes once per second, which suggests that it is this part that is down.
It's really unlucky, the device functioned only 2 minutes after unpacking, just time to see the boot and BIOS problem.
I am negotiating with the seller to have a functional PSU but I think this problem of PSU is the weak point of all CMUand CRTU-RU

Thank you all for your help.

If the led 12STB lights once per second then "maybe" the TOP200 is ok and it is activating its self-protection circuit for whatever reason, youll have to investigage

[richnormand]

Hi Richnormand  and thank you for all that information. 

I opened the CMU to display the control LED power supplies.
Of course. no LED is lit but I noticed that the LED +12V STB flashes once per second, which suggests that it is this part that is down.
It's really unlucky, the device functioned only 2 minutes after unpacking, just time to see the boot and BIOS problem.
I am negotiating with the seller to have a functional PSU but I think this problem of PSU is the weak point of all CMUand CRTU-RU

Thank you all for your help.

If the 12V standby LED flashes once a second  it might indicate the supply is trying to start and then shuts down due to a malfunction and repeats the cycle.

This could be a short on the rail or a fault in the psu itself. You could remove the psu completely and see if the 12V standby works OK by itself (no load). If it does, the problem is in one of the plugin units or the motherboard or the front screen/computer module or other stuff such as an over-temperature sensor.

You could remove each modules one by one to try to isolate which one is faulty.

But first I would remove and  operate the psu by itself and see if the 12V STB is OK without any load on it.

[Ericmax]

Hello everybody.

Did not test my CMU200 on arrival, coming back to you who seem to know well the device.
For information this is a CMU200 WCDMA FDD 1100.0008.30 sold on Ebay UK theoretically with the following options:

CMU200 WCDMA FDD
Intel CPU 256MB Memory
The Unit Has the Following Options
K21, K22, K23, K24, K65 + PCMCIA U65, U99

Following my problem PSU, so I opened the CMU and I am surprised to find so few modules. I attached a picture of the open device without PSU    
If one of you just tell me if the device is complete for the given configuration for I am a bit lost in all these options.

thank you beforehand.
wekkend good to you all.

[Pieh0]

Yes, that's complete.

The K options are software, and the U options are variations of standard modules.

U65 is obviously the PCMCIA slots, and U99 means RF1 can only take 2 watts, where as the other version can take up to 50w.

[Ericmax]

Hi Pieh0.

thanks for these informations.
it remains only to find a functional PSU me to finally have fun.
The option B41, B17 and B12 interest me if I can turn it on again.

[richnormand]

Nice job Rich, eventually youll get the faulty part. Dont forget to post your notes on the psu (ie part numbers, schematics...) as it may help others in the future!

Will do. Mostly part locations and part numbers. Don't get your hope too high for complete and detailed schematics however. I'll trace what I need for a repair and understanding the problem.

********ERRATA**********
See reply #59 and 62 later in this thread.
There should be FIVE one ohm resistors and ONE diode in the lot.
I did not change the text below so that the other responses still make sense in that context.
This is what happens when you work way past your bedtime

*************************
First: I will post part placements of two of the boards as soon as I get a chance to scan my notes.

Second: the PSU is now back in action. On board 3 there are two series of 1.0 ohm resistors in parallel. The marked one is going to a comparator. ALL the resistances were in the 100 to 200k range They looked pristine and the markings was for 1 ohm.... No sign of heating, looked perfect.
I removed a couple resistances from the second set and they, indeed, measured 1 ohm each. Guess they are measuring current draw on the 400V main storage capacitor for each of the supply rails.

Plugged back the PSU in the unit expecting some magic smoke but all the supplies were according to MasterTech's table (Thanks again ).  But apart from a few beeps the front display was dead. However I could see some activity with a strong flashlight. Must be a dead fluorescent tube or the inverter board.

Fortunately it is a TDK board (doc on google) and it was the 2.0A smd fuse that was gone. I can post info if someone needs it.

A small detail that I should have given more attention to was one missing screw on the PSU outer casing when I first took it apart. With my luck I assume I ended up with the lab unit that was reassembled from various broken parts until they go rid of it.

Many thanks to all the people that contributed to this thread. I owe a beer to MasterTech in particular....

[tautech]

But apart from a few beeps the front display was dead. However I could see some activity with a strong flashlight. Must be a dead fluorescent tube or the inverter board.

Fortunately it is a TDK board (doc on google) and it was the 2.0A smd fuse that was gone. I can post info if someone needs it.

The thread wouldn't be complete unless you did. 

These are easily overlooked, I've seen it in Philips monitors and have wondered if it's a symptom of failing caps in the backlight HV converter. 

[Pieh0]

So just to be clear, the thing that killed a piece of test gear that cost tens of thousands of dollars (brand new), with a top quality power supply from PULS, was kicked in the nuts by some 1 ohm resistors?

I've got 2 CMU200's with 1 dead psu and the other looking like it's also on the way out, so it would be nice if this is the fix .

[richnormand]

So just to be clear, the thing that killed a piece of test gear that cost tens of thousands of dollars (brand new), with a top quality power supply from PULS, was kicked in the nuts by some 1 ohm resistors?

I've got 2 CMU200's with 1 dead psu and the other looking like it's also on the way out, so it would be nice if this is the fix .

*****ERRATA, actually 5 one ohm resistors and one diode********


Actually it was 6 one ohm resistors in parallel. It was 1/6 ohm initially monitoring the current taken from the main 400V cap via the comparator IC for that rail. I guess if one failed it gets worse and the next one goes and so on, but none showed any sign of external failure. Each was over 100k in resistance when I measured them. Keep in mind the other set (bottom of pix) of 1 ohm resistor were all OK. That would limit the possibility this was a bad batch as the others are OK.

Overall it's kind of weird. Took me a while since no schematics are available and it is a pretty sophisticated PSU.

Good luck with fixing yours. If I can help, let me know.


My next issue is to fix the small fan to the front panel that is seizing up. Time to take it apart for the nth time again.

[PartialDischarge]

Thats very good news!, definitively seems like a design flaw, hope many more broken psus of the same type can be fixed with this info.

[richnormand]

My next issue is to fix the small fan to the front panel that is seizing up. Time to take it apart for the nth time again.

Dismantled the front panel again and oiled the small fan that feeds the CPU. Did the same to the main fan in the rear. All quiet now.

Here are my notes. Most power devices on two of the boards have been identified. The third one is not complete as found the issue while going through the process.
Also is a diagram of the PSU motherboard connector. And don't forget the attachment from MasterTech with all the voltages.

[valgamaa]

It's interesting that all the resistors on the PCB, other than the ones that failed, are MELF. The PCB pads also look like they are intended for MELF resistors, so the missing screw suggests you may have been the victim of a low-quality previous repair.
The 0805 (guessed size) resistors may not have the same heat dissipation capabilities, so they may well have cooked slowly, leading to the failure.
All in all a rather tricky one to fix.

[richnormand]

It's interesting that all the resistors on the PCB, other than the ones that failed, are MELF. The PCB pads also look like they are intended for MELF resistors, so the missing screw suggests you may have been the victim of a low-quality previous repair.
The 0805 (guessed size) resistors may not have the same heat dissipation capabilities, so they may well have cooked slowly, leading to the failure.
All in all a rather tricky one to fix.

The failed resistors were all MELF just like the ones at the bottom of the photo. The photo is after my repair, with what I had on hand that would fit in the space. Sorry, I should have been more precise in the description. I figured if it does not last I'll simply coil a length of wire that amounts to 1/6 ohm around the board or wire a remote resistor where space permits. So far so good as the unit has been working nonstop for two days now. But, yes, I do expect to get back in there eventually as this was a quick and dirty fix to see if that was the root issue. As stated in my previous post, I was expecting some magic smoke but it is still running.....

I also used my Seek Thermal  to check for heating but that was with the PSU operating alone  (ie. no load). I have not put the outer case back together either..... Playing the waiting game

[valgamaa]

The failed resistors were all MELF just like the ones at the bottom of the photo.

Ah, sorry, I assumed it was a 'before' picture.... with such a low resistance the resistors are unlikely to dissipate anything close to their rated power, so the repair is unlikely to need modifying.

[PartialDischarge]

The failed resistors were all MELF just like the ones at the bottom of the photo.

Ah, sorry, I assumed it was a 'before' picture.... with such a low resistance the resistors are unlikely to dissipate anything close to their rated power, so the repair is unlikely to need modifying.

Well, not so fast, resistors can get damaged by power OR voltage. Paralleling them doesnt lower the peak voltage they suffer, and in circuits like this where a cap is charged through them from mains strange things can happen with overshoot voltages especially if an inductance is somehow present... Most probably this is how those resistors failed

[Ericmax]

Hello everybody.

I received my TOP222YN. after replacing the TOP, PSU on the table I finally have my 12V STB (12,23Vdc exactly).
When the PSU is mounted in the CMU, the 12V STB LED on the motherboard flashes as before.
I removed the front Panel CMU and 12V STB LED lights normally.
When I connect the CPU board only (no keyboard, no screen, no PCMCIA) LED 12V STB flashes again.
I changed most of 820?f 6,3V on the CPU but the problem is still there.
PSU mounted in the CMU without front panel, I put b1 to GND, the CMU starts and all control power supplies LEDs are on and the fans turn.

to be continued 

[PartialDischarge]

Hello everybody.

I received my TOP222YN. after replacing the TOP, PSU on the table I finally have my 12V STB (12,23Vdc exactly).
When the PSU is mounted in the CMU, the 12V STB LED on the motherboard flashes as before.
I removed the front Panel CMU and 12V STB LED lights normally.
When I connect the CPU board only (no keyboard, no screen, no PCMCIA) LED 12V STB flashes again.
I changed most of 820?f 6,3V on the CPU but the problem is still there.
PSU mounted in the CMU without front panel, I put b1 to GND, the CMU starts and all control power supplies LEDs are on and the fans turn.

to be continued 

The 12VSTB goes directly to the CPU, see my previous post with a PDF that describes where that signal goes (exact connector pin also). You could attach a lab power supply to that exact pin, inputting 12V with a current limit to see if its drawing more than 100-200mA

[Ericmax]

Hello.

yes I have now.
I limit 250mA and U drops to 0.6 V on my lab power supply.
I thought of a chemical capacitor or SMD tantalum or diode.
I look for a component with a high temperature but would be much simpler with the thermal camera.

[PartialDischarge]

Hello.

yes I have now.
I limit 250mA and U drops to 0.6 V on my lab power supply.
I thought of a chemical capacitor or SMD tantalum or diode.
I look for a component with a high temperature but would be much simpler with the thermal camera.

theres a trick, if you dont have a thermal camera use a small amount of clear flux, it will evaporate quickly on the hot part